On 20.02.16 09:20:05, Ard Biesheuvel wrote: > On 20 February 2016 at 02:13, David Daney <ddaney.cavm@xxxxxxxxx> wrote: > > From: David Daney <david.daney@xxxxxxxxxx> > > > > v11: > > - Dropped cleanup patches for other architectures, they will be > > submitted as a separate set after more testing. > > > > - Added patch set from Ard Biesheuvel that are needed to make > > the whole thing actually work. Previously this was a > > separate set. > > > > This series is out of date, unfortunately. The EFI init code is now > (as of v4.5-rc1) shared between ARM and arm64, which means that any > changes you make must be made on both sides. This applies to the split > between parsing the EFI dt nodes and performing the actual EFI init, > but also to the early_init_dt_add_memory_arch() changes. I am happy to > have a go at this, but first I need a clear statement from whoever > maintains that area that keeping memory nodes *just* for the > annotations (and otherwise ignore them) is the way forward (Rob? > Grant?) Wasn't there the approach to check for consistency between efi tables and devicetree? Thus, DT is actually not ignored but rather checked if it is in sync with efi tables. -Robert -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
